V0.2 Important Notes: PSA: Trained mostly on the openings 'cause even the Blu-ray version was butchered...💀 Trained with Danboru tags and Natural language, real pain in the ass to manually write/edit but, better prompt adherence. 4000 Steps with 132 Images with Tags and 146 with NL Average Black clover episode animation.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/eccfcc87-bf9c-4f82-b0b6-5706be6bfe61/original=true/eccfcc87-bf9c-4f82-b0b6-5706be6bfe61.jpeg" /> A style LoRA for Anima (by extension its finetunes) by <a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs">CircleStone Labs</a>. This LoRA was trained on a curated dataset of 132+14 screenshot from the openings of Black Clover Season 1. All images were treated to remove text, and most symbols, then batch tagged with <a target="_blank" rel="ugc" href="https://huggingface.co/Mooshie/convnextv2_huge.dbv4-full">ConvNextV2 Huge Full</a> and manually tagged for fixing/adding tags. This Lora is best used at 1 strength, imo. Trigger word is very optional, but hey, its BL4CKCLOVER. Also this was trained on my chef's kiss AMD 7900XT, (under Linux with <a target="_blank" rel="ugc" href="https://github.com/ostris/ai-toolkit/">Ostris's Ai-Toolkit</a>❤️)

BLACK-CLOVER-STYLE-ANIMA

Main Settings: 4000Steps, AdamW8bit, 1Batches, [768], LR 5e-5, WeightDecay 1e-2, Rank32, Repeats 2(TAG) and 1(CAPTION), No X Axis Flip, 132 IMG (TAGS), 146 IMG (NL) [Trained with both Danbooru Tags and Natural Language. Tags weight 1.4, Caption Weight 0.6]

Files with SHA256: b4bc385141a3d1f2950cf7292e4fd68403aff1c1b3061789d996d81045982b6f